contract-compare

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted contract documents that could contain hidden instructions or adversarial content.
  • Ingestion points: Contract text is ingested via the Read tool from user-provided file paths or via pasted text blocks labeled "Version A" and "Version B" (SKILL.md, Instructions section).
  • Boundary markers: Absent. The instructions do not provide explicit delimiters or "ignore embedded instructions" warnings to the agent regarding the content of the contracts.
  • Capability inventory: The skill is configured with Read, Glob, and Grep tools and writes the comparison results to a new markdown file named CONTRACT-COMPARISON-{YYYY-MM-DD}.md.
  • Sanitization: Absent. There is no mention of sanitizing, escaping, or validating the contract text before the agent performs analysis or generates the report output.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:43 AM
Security Audit — agent-trust-hub — contract-compare